2025 AI for Good Hackathon

Dates: October 31 to November 2, 2025

AI for Good is a 48-hour hackathon that brings together students and partners, including The Link, the City of Jacksonville, and mySideWalk, to build solutions that drive social good benefits.

Join us and be a part of AI4Good, where we build AI solutions that matter, code with purpose, collaborate with peers, and innovate for real-world change.

Follow the Devpost site below for hackathon participation information, rules, judging criteria, rules, schedule, and registration details.

2025 AI4Good Devpost Site

This hackathon has ended!

Hackathon Overview

Over the course of a weekend, students from UNF, UCF, FSU, and Georgia Tech collaborated to work on unique problems impacting affordable housing. Students were provided with the tools, support, and data needed so that they can write the code to right the world.

48
Hours

29
Students

11
Teams

Participating students received mentorship from City of Jacksonville, NLP Logix, Dun & Bradstreet, CEVA Logistics, JEA, and UNF experts. Participants were expected to use their skills in the problem domain, programming, data mining, machine learning, statistical analysis, visualization, and AI techniques to solve the problem. Hackathon participants were expected to form a team with a maximum of three to work on the hackathon problems and datasets.

Social Good Problem Cases

The 2025 AI4Good hackathon aligns closely with the goals of the City of Jacksonville's State of Jax initiative that was launched to highlight Jacksonville’s most pressing challenges and opportunities. Participants were challenged to use data to drive actionable insights on the affordable housing issue in Jacksonville. By grounding hackathon work in data-driven approaches, teams can contribute recommendations that directly support informed decision-making, equitable development, and the city’s broader mission of fostering thriving, inclusive communities. Hackathon participants were provided with three problem cases and expected to solve one of them.

Problem Case #1: Predicting Housing Displacement Risk

Develop a predictive AI model to assess housing retention and displacement risk in Jacksonville compared to other cities.

Problem Case #2: City and Tract-level Affordability Indexes

Develop a city-level and tract-level affordability index using housing, transportation, walkability, and income data.

Problem Case #3: Strategic Recommendations for Affordable Housing Investment

Assess the Impact of Housing Affordability on Social Equity and Community Well-being, and Recommend Targeted Investment for Affordable Housing Preservation and Expansion.

Data Partner

Datasets relevant for the hackathon were obtained from mySidewalk through the partnership with the City of Jacksonville. Participants were provided with a primary set of datasets relevant to the issue of affordable housing. Participants were allowed to request additional data available in mySidewalk during the hackathon to enhance their analysis and AI solutions.

mySidewalk is a data platform that provides public data and intuitive tools for creating maps, reports, and visualizations to help communities, organizations, and government agencies make data-driven decisions. The platform integrates data from various trusted sources and offers templates for common needs, including demographics, housing, public health, and economic development.

Hackathon Location

The Link is an environment that fosters creativity, productivity, and human development. It is a premier coworking and event space with dedicated and floating desks, community spaces, lounges, and meeting rooms with amenities like fast Wi-Fi. The Link is equipped with state-of-the-art technology, including an immersive studio and video conferencing equipment, and hosts regular events for professionals and families, all contributing to its reputation as a community hub for work and play. The Link has facilities for showering and refreshing for event attendees.

The Link offers flexible workspace solutions and a collaborative environment perfectly suited for generating and executing innovative ideas in a short timeframe.

The Link at Ponte Vedra — A collage of The Link event spaces.

Award Sponsors

Dun & Bradstreet (D&B) provides essential business intelligence, data, and analytics, helping companies manage risk, find customers, and improve operations through services like credit risk assessment, sales and marketing insights, and supply chain management, all powered by its vast global database of businesses and its unique D-U-N-S Number. Dun & Bradstreet sponsored the first-place award.

Fidelity National Information Services, Inc. (FIS) is a multinational corporation which offers a wide range of financial products and services. FIS is the largest processing and payment FinTech company in the world. FIS sponsored the second-place award.

NLP Logix is a custom artificial intelligence and machine-learning solutions provider that focuses on automating business processes across many industries. Katie Bakewell, VP of AI Strategy sponsored the third-place award.

Schedule

4 PM: Hackathon Kickoff, problem overview, and data sources review
5 PM: Hackathon starts and continues through the night
11 AM to 12 PM, 1 PM to 2 PM, & 7 PM to 8 PM: Mentor meetings
10 AM: Stop hacking, submit the solution, and prepare for final presentation
11 AM: Evaluation of code and solutions by AI experts from UNF
1 PM: Hackathon team presentations
3:30 PM: Award ceremony

Winners

Gold Award

Horizon Team (Marion Forest, Aryan Gholinezhad, and Antony Malesevic) used an ensemble of machine learning models to compute a housing affordability score for a given zip code (or census tract) and designed an interactive dashboard you view an index of these scores.

Watch First Place Award Team presentation

Silver Award

We Love Jax team (Ehsan Ayoubi and Talia Tong) developed a AI-powered platform helps investors act faster, simulating real-time community feedback to reveal what residents value and where concerns lie.

Watch Second Place Award Team presentation

Bronze Award

CityScale team (Andrew Mahorner, Thomas Marconi, and Schmidt Joseph) computed city- and tract-level affordability risk scores using PCA components and/or configurable weights.

Watch Third Place Award Team presentation

4th Place

DojoCode team (John Butoto, Baraka Gotora, and Deogratias Nyamhanga) using historical housing data identified which areas are most at risk of housing displacement.

Watch Fourth Place Award Team presentation

Event Pictures

Flickr Album: https://flic.kr/s/aHBqjCAoYA

Hackathon Team Presentations and Award Ceremony

YouTube Playlist: 2025 AI for Good Hackathon Final Presentations Playlist on the FL-DSSG YouTube Channel

Judges

Matt Berseth

CIO and Co-Founder, NLP Logix

Ranadhir Ghosh

Director Data Science, FIS

Joshua Hicks

Affordable Housing Director, City of Jacksonville

Alan Phelan

Senior Vice President, Technology Engineering & operations, Dun & Bradstreet

Michael Vitti

SVP, Data & Decision Science, PGA TOUR

Evaluation Criteria

Each participating team will be evaluated on their model performance and presentation of the results. Winners of the hackathon will be determined based on the below evaluation factor weightings and scores received.

Presentation (20%)

Model Performance (40%)

Solution Approach (40%)

Scoring Ratings

5 = Excellent presentation on all counts

4 = Very good presentation with minor issues

3 = Pitch can be significantly improved but is satisfactory overall

2 = Pitch is incomplete and severely lacking

1 = Presenter fails to address the criteria

Mentors

Audrieanna Burgin

904ward

Nilay Chandra

Dun & Bradstreet

Indika Kahanda

UNF Computing

Mahendran Mookkiah

CEVA Logistics

Dan Richard

UNF Psychology

Laurel Wainwright

JEA

Volunteers

Danielle Graham

Grejsi Gjata

Joel David Fernandez Carrizales

Isabelle Chillakuru

Wandeir Souza

Khanh Linh Lieu

Pranavi Chintada

Markos Calderon